Large Bio Medical Databases as drivers of creativity: An analysis of the case of the Pharmaceutical Industry.

Get Complete Project Material File(s) Now! »

Some recent examples of scientific discovery: HIV and stem cells

Let us look at one example: human immunodeficiency virus (HIV) has infected more than 78 million people since 1981, it is the cause of AIDS disease and it has result in over 40 million deaths. Finding a cure to this virus is one of the biggest challenges of today’s medical research. If science is to find a preventive vaccine for HIV, which, according to Shin (2016) is still not likely to happen soon, several years of research studying how the HIV virus behaves and how it changes the human genome would be needed. These advances vary in size and relevance. For instance, research on the virus started in 1981 and the first treatment for was introduced in 1987 and it consisted on an antiretroviral (Zidovudine) that slowed down the replication of the virus in the human body and reduced the mother to child transmission during pregnancy and breast feeding. It had, however, important side effects and it didn’t prevent AIDS disease to develop (it just retarded it). Research continued and several other antiretroviral drugs were developed until in 1997 an antiretroviral therapy, which was a combination of several drugs, became the new standard treatment and caused 47 percent decline in death rates due to important delay on AIDS acquisition. Research continued and today there are more than 40 antiretroviral drugs available to control the virus and in high-income countries the standard treatment assures infected people a life expectancy that is only slightly shorter than non-infected people as death rates have shorten by more than 80%. Most people infected with HIV today in high income countries will probably never develop AIDS and the low viral load in their bodies prevents sexual transmission. The history of HIV research shows several important discoveries despite not having found a cure for the disease yet. Let us look at another relatively recent example. The discovery of the existence of stem cells was, at the time, a breakthrough discovery. Stem cells possess the innate ability to change into any kind of cell. This means they can turn into, for example, a red blood cell, a white blood cell or a muscle cell. The existence of stem cells was discovered in the 80s. After that a lot of research has been done in order to advance towards the understanding of how these stem cells work. All the discoveries that allowed us to know a bit more about stem cells are relevant. Building on those, the discovery that has been path breaking in the sense that it can by itself bring a solution to a scientific (and social) challenge was the development of a methodology that allow scientists to program those stem cells. Indeed, since 2006 we know that any cell of the body can be reprogrammed and turned into a stem cell. Moreover, once we have a stem cell, we know now how to ask it to transform in any cell we want. This is a breakthrough discovery that brings the solution to long term faced research and societal challenges. We can expect3, for instance, once clinical trials are finished, to be able to use that technique to replace damaged tissue with new cells and stem cells may be the key, as well, to be able to cure diseases such as Parkinson disease or Alzheimer. All the previous developments, however, were as well big scientific advances.

Changes in Science and Science Policy

Policy makers, when planning science policy, aim at the occurrence of these kinds of important discoveries. The objective of any policy maker and the objective and desire of the society is for science to advance which needs continuous relevant scientific developments. Sometimes to do this we will need from very large projects and some other times from everyday research that advances towards a better understanding of the problems that we are facing. How can we assure this? How to assure path breaking discoveries to occur?
Today, science is relying the more and more in big equipment. The scientific advancement has always been to some point set by the advance on instruments and the ability to do good quality research in certain areas by the access to those instruments. Today these instruments are increasingly complex and sophisticated. They often reach a cost that makes them inaccessible for single universities or laboratories (Stephan 2012). As a consequence, key instruments and other kinds of equipment are the more and more financed by public institutions and offered to the scientific community with an open logic. When provided under this logic this equipment is known ad Research Infrastructure (RI). RIs are facilities, resources and services used by the science community to conduct research. They include large scale research instruments (such as particle accelerators and telescopes), collections, depositories, public repositories (for example insect, mice or grain repositories), libraries, databases, biological archives, networks of computing facilities, research vessels, satellites and aircraft observation facilities, coastal or natural observatories, etc. The European Commission has defined them as places “to achieve excellence in highly-demanding scientific fields and simultaneously build the European Research Area (ERA)”. There is an expressed intention of achieving path breaking results and conducting disruptive science 4. We can, therefore, expect RIs to come up with big scientific advances such as the ones described before.
We have several examples, in the history of science, of discoveries that have been made possible by RIs. One example is the recent observation (2017) of the first candidate exomoon using the Hubble Space Telescope and the Kepler space telescope5. To date, astronomers have discovered a few thousands (close to 4000) of exoplanets, which are objects orbiting stars other than the Sun. A hunt for exomoons, which are bodies that orbit these distant planets, has proceeded in parallel. But these natural satellites had lingered at the limits of detection with current techniques. Another example is the discovery of the Higgs boson6 using the LHC at CERN. Scientists had theorized about sub atomic particles and how they interact with each other. For long time theories had a missing element to understand matter and this element, the Higgs boson, was theorised in the 60s. However, it was not until 2012 that this element was finally observed empirically at CERN. In order to achieve or enhance these kinds of discovery, however, there is a need for some continuity. As we explained before, the kind of very big discoveries that would make it to the news and win a Nobel prize are great. However there are a lot of smaller, but still big, everyday advances that are necessary for the advancement of science. Breakthrough are preceed by smaller discoveries but they are as well generally followed up and enriched with a flow of additional scientific outputs. RIs are expected to drive us towards all these kinds of discoveries. In order to know whether RIs are going in the right direction towards this desired goal, it is important to look at whether they are enabling not only breakthrough discoveries but also everyday knowledge creation and everyday knowledge advances. Traditionally the ability to create new knowledge have been explained as driven by creativity. Creativity (a concept better developed in Chapter 1), is the ability to produce something new and valuable. The created item may be a physical object (for instance, an invention or a piece of art) or it might be intangible (such as an idea, a musical Master piece or a scientific theory). Our interest, in this work, is creativity in science. We study creativity as the ability to produce knowledge that is new and valuable and we focus on the specific case of Research Infrastructure.

Empirical context: Trends in Science and the landscape of Large Research Infrastructures

Scientific activity has been recognised by policy makers as a fundamental driver of innovation and how both drive economic growth. However, not only science is at the origins of innovation, innovation can be crucial for the advancement of science through the development of tools, methods and instruments (Krammer, 2015; Rosenberg, 1982). Research Infrastructures lie among these elements and they have a growing importance in science with several disciplines being dependent on them. There are as well more and more large-scale projects that depend on RI. Indeed, capital-intensive research projects are developing at rapid pace in number, size and cost. In some research domains, the imperatives of the science itself requires the creation of large infrastructures and for some research domains there is simply no other way to conduct the needed experiments, observations or computations than the use of RIs.13 Furthermore, society is confronted with global-scale challenges that demand innovative, science-based solutions in areas such as health, energy, climate change-fighting and food security. All components of the research landscape are being solicited to address such global scale challenges, sometimes with efforts that are on the same vast scale as the challenges themselves14. The public investment in RIs is growing in order to answer to this call as well as to satisfy the demand of users. It is because of this growing importance of RIs for science and because of its particularities that in the past few years, there has been a growing body of literature, especially impact assessment reports, that provides with methods and tools to evaluate RIs and studies the role of RI in science (Donovan, 2011). This literature is, however, still limited and creativity is an aspect that remains disregarded in impact evaluation of RI. This reason leads us to empirically study how RIs can facilitate scientific creativity. To do so we will start by giving some descriptive elements of RIs to show their uniqueness as well as the properties that suggest that they are likely to be a proper and highlighting place for the study of creativity.

The future of science: a growing dependency on big equipment and data

Science depends the more and more on large and costly equipment such as RIs. The scientific advancement has always been to some point set by the advance on instruments and the ability to do good quality research in certain areas by the access to those instruments (Stephan 2012). Today these instruments often reach a cost that makes them inaccessible for single universities or laboratories. Because of this reason key instruments and other kind of equipment are the more and more done under the form of RIs, meaning at a large scale, by means of joint projects or financed by (inter)national funds. They also follow the more and more a logic of openness and are accessible to all researchers that require it. Additionally, RIs but also laboratories are the more and more asked to participate in large collaborations and endorse into ambitious collaboration research projects. All of this gives RIs a growing role, which is an additional reason for exploring how these kinds of objects impact scientific creativity.
Data and the associated RIs are also becoming the more and more important in science and are believed by many to become central for science in the coming years. Science has always produced data. Very early the observation of the surrounding world led to the production of data. Counting, measuring and keeping have always been important tasks in scientific activity. As an example, in the field of astronomy, the successive inventions of different tools such as the astrolabe, the quadrant, the telescope, the spectroscope have advanced towards more precise, richer data, but also more numerous and more voluminous. With recent instruments for measurement (such as satellites and large telescopes) the increase of the amount of data has grown exponentially (André, 2014). It is the same in other fields of research which have seen in recent decades their practices disrupted by the development of instruments able to produce mass data: genomics and its sequencers, climate and environmental sciences and multiple sensors at land, sea, air and space are just some examples. It is true also in social sciences, with the growing data found in the web and social networks. Together with big opportunities this mass production of data poses some technical challenges. As André (2014) explains, the rapid growth of data challenges the technological ability to store and maintain this data as well as to offer it to the scientific community. In the case of voluminous, reference data sets such as the entire human genome, there is a need for cutting edge technology, expert human resources and thus very large amounts of funding. Here again, RIs and more precisely e-infrastructures tend to play a crucial role to provide the researchers with open access to high quality databases which are very costly to maintain and require from very large storage capacity.
The society is moving towards a data-driven economy and this change could be considered as a new industrial revolution. This is true particularly for science, where the development of data collections and processing tools started a few decades ago. This is being the more and more accelerated with the development of storage and access infrastructure under an openness principle. Several scholars agree on the fact that this logic will be generalized to all the domains of great challenges. The development of these data infrastructures and its openness is believed to have the potential of meeting several societal changes in the areas of health, environment, energy and innovation. They are also the predecessor of similar societal changes in domains different from research and they are hoped to create wealth and employment15. The provision of data from research, as well as the associated processing and visualization tools, will participate obviously to the development of this economy based on the data.

Descriptive elements of RI

Because RI is a complex and multi-face object, it is difficult to reach a comprehensive definition. We will adopt the definition given by the European Commission16 on its website and on ERA and ESFRI17 documents. The term “research infrastructures” refers to “facilities, resources and services used by the science community to conduct research and foster innovation. By pooling effort and developing RIs, European countries can achieve excellence in highly-demanding scientific fields and simultaneously build the European Research Area (ERA) and Innovation Union. They include: major scientific equipment, resources such as collections, archives or scientific data, e-infrastructures such as data and computing systems, and communication networks. RIs can be single-sited (a single resource at a single location), distributed (a network of distributed resources), or virtual (the service is provided electronically)”. RIs include large scale research instruments (such as particle accelerators and telescopes), collections, depositories, public repositories (for example insect, mice or grain repositories), libraries, databases, biological archives, networks of computing facilities, research vessels, satellites and aircraft observation facilities, coastal or natural observatories, etc18. Let us study this definition in order to compile the most relevant aspects.
· From this definition we understand, firstly that there is a variety of objects that we can consider as research infrastructure since they are described as “facilities, resources and services”. This variety comes with a complexity in its study. If every single RI has a different nature and therefore different characteristics, they probably need to be studied individually. This is one of the reasons why we focus on the study of complex and dynamic processes and we use qualitative methodology.
· As we can see in the definition, RIs are “used by the science community”. RIs follow a logic of openness, the research conducted at RIs is not only performed by an internal team, but it is open to the scientific community. This makes of RI very particular places and suggests that they can be places that enable the encounter of different individuals which, as we will see later in this chapter, can enable creativity.
· The idea that RIs might be suitable for the study of creativity is also suggested by the definition of RI by itself, as they are meant to contribute to “achieve excellence in highly-demanding scientific fields”. Excellence, in science, refers to the production of knowledge of good quality that will help face societal and scientific challenges. This requires creativity and therefore the seek for excellence is a reason to expect RIs to be a suitable place for the study of creativity.
· Finally, when describing the different characteristics that RI might have and the different kinds of RIs we can find it mentions big equipment and e-infrastructures such as data. Because of the increasing relevance of these two kinds of object in science, we will study particularly these two by looking at the cases of one single sited equipment and one e-infrastructure which is virtual and consists on the provision of data.

Scientific creativity as the production of new and valuable knowledge

Similarly to other domains of knowledge, creativity in science has been described as consisting in the production of knowledge and capabilities that are new and useful (Hollingsworth, 2002) or valuable (Amabile, 1988). This definition is widely used by the literature on scientific creativity. Its main terms deserve some further reflexions. We choose Amabile’s definition and consider creativity in science as “the production of knowledge and capabilities that are new and valuable”. Knowledge needs to be new, in order to be creative, but it also needs to have some value which means that it must be considered as useful by someone. These two notions, of novelty and value, are very important for the understanding of creativity and will be present all along the thesis.

The production process of science

Science, defined as the production of knowledge and scientific activity, has been described by many scholars as a process where existing knowledge is combined in order to create new knowledge (Koestler, 1964; Simonton, 2004; Stephan, 2012). More specifically we can consider that this new knowledge is the outcome of a process of scientific inquiry, itself considered as a problem-solving activity built up by combining different elements such as concepts, perspectives or methods, in order to find answers to research questions (Simonton, 2004). This knowledge is to be transformed into a scientific publication (journal, book or other) and validated by the scientific community by means of a peer review process where the originality and relevance of the knowledge produced is evaluated (Spier, 2002; Voight and Hoogenboom, 2012). Scientific activity produces as well new capabilities such as skills, know-how and methodological development. Indeed, the production process of science is not completely linear. And in the middle of the process while using methods and techniques some new methods and techniques are developed. These, and not only the formulation of a theoretical law, are considered as a scientific output as well (Heinze et al., 2009).
Additionally the literature that studies science presents an ambiguity regarding the study of creativity, as often the words productivity and creativity are used with the same meaning and authors differ regarding their preference for one or another (Csikszentmihalyi, 2014, pp15-19) The origin of this ambiguity lies on the definitions of science by itself. What we observe with these definitions is that science, to be considered as such, requires to have at least some degree of novelty. It requires to have as well at least some degree of value. To be published, new knowledge needs to be relevant, which means that it has some worth, merit or importance to at least someone. This raises some question as the terms used to define science in general (originality and relevance) are equivalent to those used to define creative science (novelty and value). Can we consider that all science is creative? The literature review below provides us with some insights, more particularly the part dedicated to the work of Simonton (2004) where he distinguishes between big Creativity and small creativity.
Another question that has been raised by the literature regarding the definition of scientific creativity is whether the terms used to define it, that is to say, novelty and value, are compatible with each other. An important tension or even conflicting forces have been found between these two dimensions of creativity. The next subsection is dedicated to the study of that tension.

Novelty versus value: a tension between the constitutive dimensions of creativity

Despite being highly beneficial for the society, being creative in science is also tricky as the two criteria used to define creativity pull research into opposite directions. As explained by Csikszentmihalyi (2014) and Csikszentmihalyi et al (1995), there is the need to convince the guardians of the domain that the idea is creative. However, the generation of ideas that have a high degree of novelty entails going through a process in which these ideas are put at risk to be rejected because they tend to be considered as bizarre, inappropriate, unlikely or risky (Staw 1995). Creative behaviours have been defined as risky ones by several scholars (Carver & White, 1994; Keltner et al., 2003; Lee et al., 2004; Mainemelis, 2010). Novel research has a potential for a higher impact, but it also faces a higher uncertainty. Indeed, the potential impact is higher but so is the probability of a low impact (Heinze et al., 2009). Moreover, the impact of novel research might arrive with some delay, as might do the recognition of this impact by the community. Important scientific discoveries are often not well appreciated at once by the community.
It is well known that in science recognition often arrives with a certain delay. Delayed recognition is caused by different phenomena and it consists in a longer time to integrate the findings of radically novel research than incremental one (Garfield, 1977). One of the phenomena that lead to a delayed recognition is caused by the prematurity of some novel research. Research is considered premature when, because of the novelty, there is very little for other scientists to build up on. “A discovery is premature if its implications cannot be connected by a series of simple logical steps to generally accepted knowledge” (Stend, 1972). Prematurity, among other factors, results in delayed recognition. Another reason for delayed recognition consists in the resistance that the incumbent scientific paradigms show when new paradigms emerge (Kuhn, 1962, Merton 1973).

Table of contents :

Chapter 1: Context and Theoretical Background
1 Introduction
2 Empirical context: Trends in Science and the landscape of Large Research
2.1 The future of science: a growing dependency on big equipment and data
2.2 Descriptive elements of RI
3 A selective literature review on the notion of creativity
3.1 Creativity in science
3.1.1 Scientific creativity as the production of new and valuable knowledge
3.1.2 Combining knowledge, the work of Arthur Koestler and the mechanism of bisociation
3.1.3 From genius to chance: the works of Dean Keith Simonton
3.1.4 The impact of the organizational environment: the works of Teresa Amabile
3.2 Creativity as a collective process: contributions of management literature
3.2.1 The creative process: perspectives from the study of innovation
3.2.2 The role of organizational learning
3.2.3 The role of communities
3.3 Towards a conceptual framework
3.3.1 Synthesis of the most relevant points of the literature
3.3.2 Positioning the thesis along the dimensions of creativity
3.3.3 Building blocks of the collective science creation process
4 Epistemological and methodological foundations
4.1 Epistemological foundations
4.2 Epistemological paradigms
4.2.1 Critical realism
4.3 The reasoning model
4.4 Research methodology
4.4.1 Qualitative methodology
4.4.2 The case study
4.4.3 Methodological development and quantitative analysis
5 Concluding remarks
Chapter 2: Large Instruments as facilitators of users’ creative process: The role of organizational factors, collaborations and communities in the case of a European Synchrotron
1 Introduction
2 Theoretical Framework
2.1 Favourable factors for creativity at the synchrotron
2.1.1 Scientific autonomy and leadership
2.1.2 Scientific diversity
2.1.3 The role of communities
2.1.4 Summary of factors and objectives for the case study
2.2 The creative mechanisms: combining complementary pieces of knowledge through effective communications and interactions
2.3 Creative outcomes: The synchrotron contribution to users’ creative results
2.3.1 Knowledge creation
2.3.2 Innovation
2.3.3 Quality and impact
2.4 Conceptual framework and Research gap
3 Empirical analysis: A qualitative case study about the synchrotron SOLEIL
3.1 Context and presentation of the case: The EvaRIO project and the Synchrotron SOLEIL
3.2 Empirical analysis
3.2.1 Why choosing a case study methodology?
3.2.2 The data
3.2.3 Method of analysis
3.3 Analysis and results
3.3.1 Favourable factors to creativity
3.3.2 The creative mechanisms
3.3.3 Synchrotron contribution to creativity: knowledge, technology and communities
3.3.4 Quality and impact
4 Discussion of the results
4.1 Favourable factors and mechanisms for creativity at the synchrotron
4.1.1 Beamline scientists
4.1.2 Variety
4.1.3 The role of communities
4.2 The creative mechanisms
4.2.1 Effective communication
4.2.2 From communication to partnerships
4.2.3 Difference between beamlines
4.3 Creative outcomes
4.3.1 Novelty and value
4.3.2 Unexpected result
4.4 Summary of findings
4.4.1 Possible limits to creativity
5 Conclusion
Chapter 3: Large Bio Medical Databases as drivers of creativity: An analysis of the case of the Pharmaceutical Industry.
1 Introduction
2 Theoretical framework
2.1 Findings on scientific creativity: the importance of variety
2.2 Insights on knowledge distance and variety
2.3 Insights on the role of explicit knowledge on creativity
2.4 Research gap
3 Presentation of the case study, EBI databases and methodological approach .
3.1 Description of the EBI databases
3.2 Research Strategy
3.3 Choice of informers
3.4 Interviews
4 Analysis and results
4.1 Analysis of transcript (first wave of interviews)
4.2 Analysis of transcripts (2nd wave of interviews)
4.3 Discussion of results
4.4 Serendipity
5 Conclusion and perspectives
Chapter 4: Measuring RIs’ impact on scientific creativity. The case of the synchrotron
1 Introduction
2 Measuring scientific creativity
2.1 Finding an operational definition of creativity
2.2 Metrics for the study of creativity in science
2.2.1 Traditional focus on value through the measurement of impact
2.2.2 Focusing on the measurement of novelty
3 An empirical exercise on synchrotron beamlines
3.1 Methodology
3.1.1 Part 1: impact analysis
3.1.2 Part 2: novelty analysis
3.2 The data
3.2.1 Beamline papers
3.2.2 Journal universe and scientific categories
3.2.3 WoS notices
3.3 Analysis and results
3.3.1 Scientific field and (relative) impact of beamline papers
3.3.2 (Un-)commonness of journal citations and citation combination in beamline papers 240
3.4 Discussion of results
4 Conclusion
Appendix Chapter 4
General Conclusion
1 General Outlook
2 Main results
2.1 Supportive conditions and mechanisms for creativity
2.2 Types of creative outcome
2.3 Methodology for evaluation of impact on creativity
3 Policy and managerial implications
3.1 Managerial implications
3.2 Policy Implications
4 Limits and perspectives
5 Concluding Remarks
References

GET THE COMPLETE PROJECT